AI Breakthrough: Routine Blood Tests Predict Spinal Cord Injury Outcomes

AI breakthrough finds life-saving insights in everyday bloodwork

The research team utilized advanced analytics and machine learning, a type of artificial intelligence, to assess whether routine blood tests could serve as early warning signs for spinal cord injury patient outcomes. More than 20 million people worldwide were affected by spinal cord injury in 2019, with 930,000 new cases each year, according to the World Health Organization. Traumatic spinal cord injury often requires intensive care and is characterized by variable clinical presentations and recovery trajectories, complicating diagnosis and prognosis, especially in emergency departments and intensive care units. "Routine blood tests could offer doctors important and affordable information to help predict risk of death, the presence of an injury and how severe it might be," said Dr. Abel Torres Espín, a professor in Waterloo's School of Public Health Sciences. The researchers sampled hospital data from more than 2,600 patients in the U.S. They used machine learning to analyze millions of data points and discover hidden patterns in common blood measurements, such as electrolytes and immune cells, taken during the first three weeks after a spinal cord injury. They found that these patterns could help forecast recovery and injury severity, even without early neurological exams, which are not always reliable as they depend on a patient's responsiveness. "While a single biomarker measured at a single time point can have predictive power, the broader story lies in multiple biomarkers and the changes they show over time," said Dr. Marzieh Mussavi Rizi, a postdoctoral scholar in Torres Espín's lab at Waterloo. The models, which do not rely on early neurological assessment, were accurate in predicting mortality and the severity of injury as early as one to three days after admission to the hospital, compared to standard non-specific severity measures that are often performed during the first day of arrival to intensive care. The research also found that accuracy increased over time as more blood tests became available. Although other measures, such as MRI and fluid omics-based biomarkers, can also provide objective data, they are not always readily accessible across medical settings. Routine blood tests, on the other hand, are economical, easy to obtain, and available in every hospital. "Prediction of injury severity in the first days is clinically relevant for decision-making, yet it is a challenging task through neurological assessment alone," Torres Espín said. "We show the potential to predict whether an injury is motor complete or incomplete with routine blood data early after injury, and an increase in prediction performance as time progresses. "This foundational work can open new possibilities in clinical practice, allowing for better-informed decisions about treatment priorities and resource allocation in critical care settings for many physical injuries." The study, Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury, was published in Nature's NPJ Digital Medicine Magazine.

Blood Tests Could Predict Spinal Cord Injury Severity and Survival - Neuroscience News

Summary: A new study shows that routine hospital blood tests could help predict spinal cord injury severity and survival chances. Researchers used machine learning to analyze data from thousands of patients and found that patterns in blood markers, such as electrolytes and immune cells, forecasted recovery outcomes as early as one to three days after admission. Unlike neurological exams, which depend on patient responsiveness, this method offers objective and reliable insights. The findings could improve emergency care and resource allocation for spinal cord injuries worldwide. Routine blood samples, such as those taken daily at any hospital and tracked over time, could help predict the severity of an injury and even provide insights into mortality after spinal cord damage, according to a recent University of Waterloo study. The research team utilized advanced analytics and machine learning, a type of artificial intelligence, to assess whether routine blood tests could serve as early warning signs for spinal cord injury patient outcomes. More than 20 million people worldwide were affected by spinal cord injury in 2019, with 930,000 new cases each year, according to the World Health Organization. Traumatic spinal cord injury often requires intensive care and is characterized by variable clinical presentations and recovery trajectories, complicating diagnosis and prognosis, especially in emergency departments and intensive care units. "Routine blood tests could offer doctors important and affordable information to help predict risk of death, the presence of an injury and how severe it might be," said Dr. Abel Torres Espín, a professor in Waterloo's School of Public Health Sciences. The researchers sampled hospital data from more than 2,600 patients in the U.S. They used machine learning to analyze millions of data points and discover hidden patterns in common blood measurements, such as electrolytes and immune cells, taken during the first three weeks after a spinal cord injury. They found that these patterns could help forecast recovery and injury severity, even without early neurological exams, which are not always reliable as they depend on a patient's responsiveness. "While a single biomarker measured at a single time point can have predictive power, the broader story lies in multiple biomarkers and the changes they show over time," said Dr. Marzieh Mussavi Rizi, a postdoctoral scholar in Torres Espín's lab at Waterloo. The models, which do not rely on early neurological assessment, were accurate in predicting mortality and the severity of injury as early as one to three days after admission to the hospital, compared to standard non-specific severity measures that are often performed during the first day of arrival to intensive care. The research also found that accuracy increased over time as more blood tests became available. Although other measures, such as MRI and fluid omics-based biomarkers, can also provide objective data, they are not always readily accessible across medical settings. Routine blood tests, on the other hand, are economical, easy to obtain, and available in every hospital. "Prediction of injury severity in the first days is clinically relevant for decision-making, yet it is a challenging task through neurological assessment alone," Torres Espín said. "We show the potential to predict whether an injury is motor complete or incomplete with routine blood data early after injury, and an increase in prediction performance as time progresses. "This foundational work can open new possibilities in clinical practice, allowing for better-informed decisions about treatment priorities and resource allocation in critical care settings for many physical injuries." Author: Ryon Jones Source: University of Waterloo Contact: Ryon Jones - University of Waterloo Image: The image is credited to Neuroscience News Original Research: Open access. "Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury" by Abel Torres Espín et al. npj Digital Medicine Abstract Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury Routinely collected blood tests can reflect underlying pathophysiological processes. We demonstrate that the dynamics of routinely collected blood tests hold prediction validity in acute Spinal Cord Injury (SCI). Using MIMIC data (n = 2615) for modeling and TRACK-SCI study data (n = 137) for validation, we identified multiple trajectories for common blood markers. We developed machine learning models for the dynamic prediction of in-hospital mortality, SCI occurrence in spine trauma patients, and SCI severity (motor complete vs. incomplete). The in-hospital mortality model achieved an out-of-train ROC-AUC of 0.79 [0.77-0.81] day one post-injury, improving to 0.89 [0.88-0.89] by day 21. For detecting the presence of SCI after spine trauma, the highest ROC-AUC was 0.71 [0.69-0.72] achieved by day 21. By day seven, the ROC-AUC for SCI severity was 0.81 [0.77-0.85]. Our full models outperformed the severity score SAPS II following seven days of hospitalization.

Machine learning unlocks blood test secrets for spinal cord injury

University of WaterlooSep 23 2025 Routine blood samples, such as those taken daily at any hospital and tracked over time, could help predict the severity of an injury and even provide insights into mortality after spinal cord damage, according to a recent University of Waterloo study. The research team utilized advanced analytics and machine learning, a type of artificial intelligence, to assess whether routine blood tests could serve as early warning signs for spinal cord injury patient outcomes. More than 20 million people worldwide were affected by spinal cord injury in 2019, with 930,000 new cases each year, according to the World Health Organization. Traumatic spinal cord injury often requires intensive care and is characterized by variable clinical presentations and recovery trajectories, complicating diagnosis and prognosis, especially in emergency departments and intensive care units. Routine blood tests could offer doctors important and affordable information to help predict risk of death, the presence of an injury and how severe it might be." Dr. Abel Torres Espín, Professor, School of Public Health Sciences, University of Waterloo The researchers sampled hospital data from more than 2,600 patients in the U.S. They used machine learning to analyze millions of data points and discover hidden patterns in common blood measurements, such as electrolytes and immune cells, taken during the first three weeks after a spinal cord injury. They found that these patterns could help forecast recovery and injury severity, even without early neurological exams, which are not always reliable as they depend on a patient's responsiveness. "While a single biomarker measured at a single time point can have predictive power, the broader story lies in multiple biomarkers and the changes they show over time," said Dr. Marzieh Mussavi Rizi, a postdoctoral scholar in Torres Espín's lab at Waterloo. The models, which do not rely on early neurological assessment, were accurate in predicting mortality and the severity of injury as early as one to three days after admission to the hospital, compared to standard non-specific severity measures that are often performed during the first day of arrival to intensive care. The research also found that accuracy increased over time as more blood tests became available. Although other measures, such as MRI and fluid omics-based biomarkers, can also provide objective data, they are not always readily accessible across medical settings. Routine blood tests, on the other hand, are economical, easy to obtain, and available in every hospital. "Prediction of injury severity in the first days is clinically relevant for decision-making, yet it is a challenging task through neurological assessment alone," Torres Espín said. "We show the potential to predict whether an injury is motor complete or incomplete with routine blood data early after injury, and an increase in prediction performance as time progresses. "This foundational work can open new possibilities in clinical practice, allowing for better-informed decisions about treatment priorities and resource allocation in critical care settings for many physical injuries." University of Waterloo Journal reference: Rizi, M. M., et al. (2025) Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury. Npj Digital Medicine. doi.org/10.1038/s41746-025-01782-0.